Deciding authorship beforehand: Does this actually work?

[futureapppsy2]

I’ve repeatedly seen a lot of people say that authorship order is best determined at the start of a manuscript process, but over 80+ articles with lots of different people, I’ve almost never seen it work. What people do in reality is often very different than what they plan to do, not out of any sort of malice, but just because availability and interest is fluid. I’ve actually seen it backfire a fair bit. For example, I have one article where I did 98% of the work—IRB, funding the study, all the data analysis, all the writing, even the proofs—but I’m second author because that’s what my co-author and I had agreed on at the outset (I just wanted it published and didn't really care, honestly). In other situations, I’ve seen authorship change at the end because people did more or less work than they planned when the order was decided at the beginning and people getting absolutely furious about that, because they felt duped or cheated either way. I’ve always found that it works better to keep order and tasks fluid and then reach consensus on order at the end, so things keep moving along and everyone does what they actually have the time/interest for.


So, I honestly don’t get why this is a common best practice if I’ve almost never seen it work. Am I missing something here?

---

Members don't see this ad.

 

[tr]

I don't think the author order discussed at the outset is set in stone, but it's helpful to get an idea of expectations and ferret out if there are any clashing ones before the bulk of the writing is done. I agree that there are always going to be changes and the proportion of work done by each author by the end may be different from what was expected in the beginning, but it's easier to have a discussion about changes if everyone was on the same page to start with, rather than discovering at the very end of the process that, e.g., two different people both expected to be first author. I usually put a 'tentative author order' at the top of the first outline I send around and invite further discussion at that point.

 

[Ollie123]

I think I have honestly only had two articles where we determined it in advance (with one more currently in the hopper....you know which one!).

Generally speaking, we just discuss who is going to lead it. First author generally does 90% of the writing, last author usually is the senior mentor or the "owner" of the data, in the middle is a scattering of co-investigators, post-docs, students, etc. who contributed to the project but relatively little to the writing itself. Occasional exceptions (e.g. when the analytics are complicated and there is a separate statistician not first or last author).

Generally speaking though, no one really cares beyond first/last author position. I think it's more common for students to care and make a big deal about it, but generally speaking any middle author position is viewed as about equal whether it is 2nd or 52nd. Admittedly I am also used to large team science projects (have a paper under review right now with ~40 authors).

 

[MAClinician]

Admittedly I am also used to large team science projects (have a paper under review right now with ~40 authors).

Just the byline could take up half a page or more?

 

[albatross_at_crossroads]

Just the byline could take up half a page or more?

I've recently read a study (genetics applied in mental health) where just listing the authors took the first page, plus a couple of rows on the second page. Can send the reference if you are interested 🙂

On topic - Indeed, authorship should be about major writer and senior researcher of the lab who owns the data and/or funded its collection.

However, as @futureapppsy2 said, things in real life sometimes take a weird turn. In the past I wrote and submitted an article to several places with a mentor, and it got rejected (since it was more mediocre and they were aiming really high) and the mentor decided to re-write part of it, add some new experiments and keep submitting. It ended up being accepted but because I was out of academia at that point, they left me completely out and replaced me with another student. It doesn't really bother me because I learned through experience with other researchers that that particular mentor wasn't exactly the best when it came to how to do research properly. But the program they are in are more interested in how much they publish and don't offer a lot of guidelines on how authorship and contributions should work.

If you can, best way to do it is try to establish some "tentative" authorship and be clear on responsibilities for each author, with the caveat that if workloads change, there would be some amendments. This is pretty ideal and I'm not sure how easy it is to implement in real life though.

 

[Ollie123]

Just the byline could take up half a page or more?

Sometimes. Different journals handle it differently. It sometimes gets abbreviated to "The XYZ Consortium" and actual names are in a supplement.

This was a methodology paper developing a new methodology checklist for a specific type of study that was assembled by a large team. Psychologists (esp those who don't do much interdisciplinary work) are sometimes shocked but author lists like that are super common in many fields. Scroll through Nature Genetics at some point if you want some examples as you will probably see at least a few in every issue.

 

[R. Matey]

Determining and negotiating authorship

No one uses this? Our lab did.

 

[PsyDuck90]

Determining and negotiating authorship

No one uses this? Our lab did.

My lab uses this too.

 

[futureapppsy2]

Sometimes. Different journals handle it differently. It sometimes gets abbreviated to "The XYZ Consortium" and actual names are in a supplement.

This was a methodology paper developing a new methodology checklist for a specific type of study that was assembled by a large team. Psychologists (esp those who don't do much interdisciplinary work) are sometimes shocked but author lists like that are super common in many fields. Scroll through Nature Genetics at some point if you want some examples as you will probably see at least a few in every issue.

I have one (maybe two?) articles where I’m part of the “and XYZ consortium”, but I don’t consider them to be my publications.

 

[EmotRegulation]

I agree with others that discussing authorship *and roles* ahead of time makes sense, but it can't be set in stone. I've had students agree to take first author on things but then never finish the project, and it ends up getting passed on to someone else who has to do/redo analyses and write the brunt of the paper....in these instances, authorship order is likely to change. But I also think it makes sense to outline who is going to do what on the front end, so that people have a sense of what they are agreeing to--that initial discussion is kind of like consent. But just like clients sometimes require a violation of consent or just drop the hell out of therapy, things change over time. 🙂

 

[member978]

Throughout my graduate training, authorship decisions were generally a lot more fluid other than who would lead the paper. I also saw things change over time, interests (and time/availability) wax and wane and so authorship changes would follow. Writing in graduate school was also just as much about learning how to write as anything else, and so sometimes papers would take some time to produce.

In my current role at an academic medical center, I am part of an extremely efficient and productive team which uses a formal writing contract when deciding on authorship. Generally the first author comes up with the concept, assembles the team based on interest, assigns tasks and everyone collaborates. Each author generally has a specific task to accomplish (e.g., write the intro, do the analyses, make tables) and a timeline for completion. I've seen this process work extremely efficiently, with an entire manuscript coming together from nothing to submission way faster than I was ever used to in graduate school. In part it works because everyone is far along in their training and can whip up an intro in their area of expertise in a matter of hours, rather than weeks of writing and revising.

My personal opinion is probably somewhere in the middle, I like the idea of the structure and set expectations that comes from having an up-front discussion. I have seen authorship problems where someone will end up with authorship on a manuscript after doing essentially zero work on the paper, conceptually or practically, and I think having defined tasks and contributions might help curb that. That being said, I also think a bit of flexibility is required as things can easily change and authorship can reflect that.